ArticlePDF Available

Abstract and Figures

FexPt1-x nanoparticles prepared by organometallic synthesis and gas-phase condensation were structurally and magnetically characterised. The effective spin magnetic moments at both the Fe and Pt sites are reduced with respect to the moments in the corresponding bulk material by up to 20% and further decrease with decreasing particle size at the Fe sites. The ratio of orbital-to-effective-spin magnetic moment at the Fe sites increases from 2.1% for 6 nm particles to 3.4% for 3.4 nm particles due to the break of symmetry at the surface. A lowering of the crystal symmetry after the transformation to the chemically ordered L10 state yields a ≈ 9% and is accompanied by an enhancement of the coercive field at 15 K from (36±5) mT to (292±8) mT indicating an increase of the anisotropy.
Content may be subject to copyright.
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Modern Physics Letters B, Vol. 21, No. 18 (2007) 1111–1131
c
World Scientific Publishing Company
MAGNETISM AT THE NANOSCALE: THE CASE OF FePt
C. ANTONIAK and M. FARLE
Fachbereich Physik and Center for Nanointegration Duisburg-Essen (CeNIDE),
Universit¨at Duisburg-Essen, Lotharstr. 1,
47048 Duisburg, Germany
FexPt1xnanoparticles prepared by organometallic synthesis and gas-phase conden-
sation were structurally and magnetically characterised. The effective spin magnetic
moments at both the Fe and Pt sites are reduced with respect to the moments in the
corresponding bulk material by up to 20% and further decrease with decreasing particle
size at the Fe sites. The ratio of orbital-to-effective-spin magnetic moment µleff
Sat the
Fe sites increases from 2.1% for 6 nm particles to 3.4% for 3.4 nm particles due to the
break of symmetry at the surface. A lowering of the crystal symmetry after the trans-
formation to the chemically ordered L10state yields a µleff
S9% and is accompanied
by an enhancement of the coercive field at 15 K from (36 ±5) mT to (292 ±8) mT
indicating an increase of the anisotropy.
Keywords: Nanoparticles; anisotropy; magnetic moments.
1. Introduction
Due to finite size effects,1different crystal structures and large surface contribu-
tions, the properties of nanoparticles may differ significantly from those of the cor-
responding bulk material and have been the subject of intense research activities
in the last years.
About 20% of the atoms of a spherical nanoparticle with a diameter of 6 nm
are surface atoms, making ensembles of nanoparticles suitable for studying surface
effects or interface induced phenomena in core-shell nanoparticles. The latter appear
quite often: Nanoparticles containing ignoble metals which are usually characterised
after transport through air after synthesis, oxidise and form metal-metal oxide core-
shell particles like e.g. in the case of Co or FePt. The oxide shell may be as thin
as one or two atomic surface layers, but also the whole nanoparticle may be fully
oxidised depending on the time of exposure to air.2The surface anisotropy and
the formation of antiferromagnetic oxides3like CoO or α-Fe2O3may lead to a
complex spin structure. In that case, the direction of the magnetic moments are
forced into different directions by the different anisotropies like volume, surface,
interface anisotropy and the exchange interaction between the antiferromagnetic
(AFM) oxide and the ferromagnetic (FM) metal.4
Here, we focus on pure metallic FePt nanoparticles with contributions of surface
and volume anisotropy only. Such particles can be investigated only in situ after a
1111
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1112 C. Antoniak & M. Farle
soft hydrogen plasma removal of oxides and ligands which are present due to the
transport through air or the synthesis method. Ion bombardment to remove the
surface contaminations results in modifications of the shape and topography of the
particles.
Since the exchange energy in particles up to 6 nm of FexPt1xalloys with
exchange lengths of 20-40 nm is assumed to be larger than the surface anisotropy,
the particles are single domain particles and each particle can be treated as a
single macrospin. One should note, however that the temperature dependence of
the magnetisation of particles with less than 8 nm diameter may show significant
differences from the Bloch-like bulk behavior as shown for example in the case of
FePt3nanocubes.5
The FexPt1xsystems form alloys in a wide range of concentration as can be
seen in the phase diagram (Fig. 1). Fe-rich compositions with an Fe content above
x= 0.75 form a bcc structure below 900C, all other compositions crystallise in face-
centred structures. There are three chemically ordered phases in the thermodynamic
equilibrium states: ferromagnetic Fe3Pt and antiferromagnetic FePt3, both of them
having a fcc structure with an L12symmetry, and the ferromagnetic FePt with
a fct structure and an L10symmetry. The formation of the chemically ordered
phase is driven by volume diffusion and is kinetically suppressed. The disorder-
order transformation in FePt occurs at temperatures between 500C and 600C for
FePt thin films9, 10 and nanoparticles.11
600
800
1000
1200
1400
1600
1800
temperature, T [°C]
20 40 60 80 1000
Fe content, [at%]x
melted
ä-Fe
Fe Pt
3
FePt
FePt3
fcc, chemically disordered (A1)
á-Fe
L10
1519°C
1538°C
1394°C
912°C
700°C
magnet.
transf.
1300°C
1350°C
730°C
L12
L12
Fig. 1. Phase digram of FexPt1xalloys.6–8
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1113
In the phase with L10symmetry, the FePt alloy consists of alternating layers
of Fe and Pt along a h100idirection yielding a tetragonal distortion of the other-
wise cubic lattice. The deviation from a cubic symmetry together with the large
spin-orbit coupling (SOC) induced by the Pt atoms leads to the large magnetic
anisotropy of K= 6 ×106J/m3reported in the literature.11, 12 Materials with a
high magnetic anisotropy like FePt are needed to overcome the so-called superpara-
magnetic limit, i.e. the instability of the magnetic moment of a nanoparticle due to
thermally activated fluctuations, for the possible use as new high-density storage
media.13, 14
In this article, we give a short introduction to the sources of magnetic
anisotropies and the blocking behavior of nanoparticle systems in general before
we turn to the discussion of FexPt1xnanoparticles. We present recent results
about
(i) how to achieve the chemically ordered phase in FePt nanoparticles using dif-
ferent preparation methods, i.e. condensation in the gas-phase or using a wet-
chemical approach,
(ii) the alloying and lattice expansion in FexPt1xnanoparticles analysed by
means of layer-resolved high-resolution transmission electron microscopy (HR
TEM) and the extended X-ray absorption fine structure (EXAFS), and about
(iii) the element-specific magnetic moments of oxide-free nanoparticles and their
dependence on size and crystal structure.
2. Magnetic Anisotropy
The magnetic anisotropy energy (MAE) is the energy needed to turn the magneti-
sation from one of its easy directions into a hard direction of the magnetisation.
Easy direction denotes a direction with minimum, hard direction with maximum
ground state energy of the system. Without additional magnetic fields, the mag-
netic anisotropy determines the direction of the magnetisation in the crystal lattice.
There are two competitive contributions to the MAE: magnetic dipolar interactions
and the SOC. The exchange interaction should not influence the MAE, since it is
isotropic within the Heisenberg model. The anisotropy induced by the SOC mainly
depends on the anisotropy of the orbital magnetic moment ∆µland the SOC con-
stant ξ. In first approximation by treating the SOC in a second order perturbation
theory it can be written as:15
EA=αξ
4µB
µl.(1)
From first-principle calculations16–18 it is known, that αis a function of ∆µland
depends on the band structure of the material.
The second source of the MAE is the magnetic dipole-dipole interaction. A
magnetic moment ~µidescribed by a magnetic dipole that interacts with a magnetic
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1114 C. Antoniak & M. Farle
dipole ~µjat the distance ~rij has the energy
Edip =µ0
4π"~µi·~µj
|~rij |33(~rij ·~µi)(~rij ·~µj)
|~rij |5#.(2)
Since the positions of the magnetic moments and hence ~rij are linked to the crys-
tallographic axes, the magnetic dipolar interaction is anisotropic.
The magnetic dipolar interaction is the origin of poles at the surface of finite
samples that induce magnetic stray fields. The energy of a sample with a magneti-
sation ~
Min its self-induced stray field ~
Hscan be written as
Es=1
2µ0Z~
M·~
HsdV=1
2µ0Z~
M·ˆ
N~
MdV . (3)
The demagnetising tensor ˆ
Nwith its trace tr[ ˆ
N] = 1 and consequently the energy
Esis determined by the shape of the sample and leads to an anisotropy that is called
shape anisotropy. In the case of a rotational ellipsoid, ˆ
Ncan be diagonalised and the
entries Nxx,Nyy ,Nzz can be calculated exactly. For spherical samples, the shape
anisotropy vanishes due to the rotational invariance (Nxx =Nyy =Nzz = 1/3). For
a two-dimensional infinite film in the (x,y) plane, the tensor can be diagonalised
with the entries on the diagonal Nxx =Nyy = 0 and Nzz = 1. Similarly to the
latter case, the shape anisotropy of a two-dimensional dipolar coupled ensemble of
spherical nanoparticles can be described by the following equation for the energy
density Fs:
Fs=1
2fµ0|~
M|2cos2θ , (4)
where fwith 0 < f < 1 denotes a filling factor.19
Especially in nanoparticles with a large surface-to-volume ratio, the break of
symmetry at the surface leads to the so-called surface anisotropy.20, 21 In Fig. 2
the surface of a truncated octahedron is shown,22 which is the equilibrium shape of
fcc nanoparticles according to Wulff’s theorem.23 This octahedron consists of eight
(111) and six (100) facets and edges between (111)–(111) and (100)–(111) facets.
The atoms at the edges and the different facets have different coordination in terms
of the number and the direction of the connection vectors to nearest neighbour
atoms. This leads to different surface and step anisotropies and if small deviations
from the perfect octahedral shape occur, all the contributions of anisotropies are
summed up to an effective anisotropy density Keff , that is assumed to be uniaxial
in a rough approximation.
2.1. Blocking behavior of fine particle systems
If the thermal energy kBTis larger than the energy barrier ∆E=Keff Vcaused by
the effective anisotropy, the direction of the magnetisation is not fixed to one easy
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1115
(100) facet
8 nearest neighbours (n.n)
(111) facet
7 n.n
apex
6 n.n
(100)-(111) edge
7 n.n
(111)-(111) edge
7 n.n
edge axis
edge axis
(111) facet
(100) facet
Fig. 2. The different atomic positions at the surface of a perfect truncated octahedron with eight
(111) and six (100) facets. Adapted from the publication by Jamet et al.22
direction of the magnetisation, but fluctuates. The fluctuation time is described by
an Arrhenius law
τ=τ0exp E
kBT=τ0exp Keff V
kBT,(5)
with the characteristic time τ0in the range of 1012-109s for nanoparticles.24, 25
If the time window of the measurement is larger than τ, a sample of ferromag-
netic nanoparticles reacts like a paramagnet even below the Curie temperature, the
sample is in the superparamagnetic regime. If the time window is smaller than τ,
the fluctuations are not measurable and the sample is ferromagnetically blocked.
The temperature limit between the superparamagnetic and the ferromagnetically
blocked regime is the blocking temperature TBthat is not a material constant, but
depends on the method of the measurement.
Above TB, the magnetisation as a function of temperature and magnetic field
Bof an ensemble of identical nanoparticles follows the Langevin equation for para-
magnets
M=MSL[x] = MScothx1
x, x =MSV B
kBT,(6)
where MSdenotes the saturation magnetisation. In the case of small magnetic
fields, i.e. x1, the cotangent hyperbolicus can be expanded into a Taylor series
and approximation of Eq. (6) leads to
M=MS
x
3.(7)
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1116 C. Antoniak & M. Farle
Keff = 1 10 J/m´5 3
Keff = 10 10 J/m´5 3
Keff = 5 10 J/m´53
050 100 150 200 250 300 350
temperature, [K]T
TB1
TB5
TB10
magnetisation, M [arb. units]
Fig. 3. Calculated ZFC magnetisation curves for three different values of the effective anisotropy
using Eq. (9), a log-normal volume distribution with a variance of σ= 0.45 around the mean
volume of a sphere with a diameter d= 4nm,τ0= 1012 s, τ= 102s. The dashed lines show the
magnetisation curve for identical nanoparticles (σ= 0).
At temperatures below TB, the magnetisation depends on the history of the sample
that is to say if the sample was cooled in an external magnetic field (field cooled,
FC) or without any field applied (zero-field cooled, ZFC). The magnetisation for
these two cases, Mbl
F C or Mbl
ZF C , can be written as26, 27
Mbl
F C =M2
SV B
3kBT;Mbl
ZF C =M2
SB
3Keff
.(8)
Since a real sample does not consist of identical nanoparticles, but particles with
a size distribution and hence, a distribution of blocking temperatures, the effect of
the size distribution should be added to Eq. (8). The temperature dependence of
the magnetisation of an ensemble of particles with a volume distribution D(V) is
the sum over the temperature-dependent contributions of the superparamagnetic
particles and the ferromagnetically blocked particles and can be described by26, 28
M=M2
SB
3kBT
1
NZVL
0
V2D(V)dV+M2
SB
3Keff Z
VL
V D(V)dV(9)
with the normalisation constant N=R
0V D(V)dVand the limit volume VL=
kBT
Keff ln τ
τ0between superparamagnetic and ferromagnetically blocked particles. In
Fig. 3 calculated ZFC magnetisation curves are shown for three different values
of Keff . The time window was set to τ= 102s, as it is the case for magnetom-
etry by means of a superconducting quantum interference device (SQUID). The
volume distribution is assumed to be log-normal around the volume of a sphere
with a diameter d= 4 nm and a variance σ= 0.45. The magnetisation curves
for an ensemble of identical particles (σ= 0) are shown as dashed lines and the
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1117
corresponding blocking temperatures are labelled TB1,TB5,TB10 . Note that for
finite size distributions (σ6= 0), the mean blocking temperature is smaller than the
temperature at which the ZFC magnetisation is maximum.
Calculations of the ZFC magnetisation curve according to Eq. (9) and fitting
to experimental data, is a way to extract the effective magnetic anisotropy for
fine particle systems with a negligible dipolar coupling between the particles as
in the case for Fe0.70Pt0.30 nanoparticles with a mean diameter of about 2.3 nm
and an interparticle spacing of 3.5 nm.28 Figure 4 shows the experimental data
of the ZFC magnetisation of this system measured with a SQUID and ferromag-
netic resonance (FMR). In the FMR, the intensity of the microwave absorption
is proportional to the magnetic moment. The time windows of the two methods
differ, yielding different blocking temperatures. The calculated curves were fitted
to the experimental data using Keff = 8.4×105J/m3at the SQUID blocking tem-
perature TSQUID
B= 23 K and τ0= 1.7×1012 s. From temperature dependent
FMR measurements, the temperature dependence of the effective anisotropy was
determined28 and taken into account in the simulations. One may note, that due
to thermal fluctuations, this method overestimates the temperature dependence29
of the anisotropy, yielding a vanishing anisotropy below the Curie temperature.
0100 200 300 400
temperature, [K]T
0
0.1
0.2
0.3
0.4
0.5
FMR intensity [arb. units]
SQUID
FMR
total magnetic moment, [10 Am ]ìtot
-6 2
TB
FMR
TB
SQUID
Fig. 4. ZFC magnetisation curves measured with a SQUID magnetometer and using the FMR
technique (open symbols). Solid lines represent calculated data according to Eq. (9).
3. Synthesis
Mainly, two methods of FePt nanoparticles preparation exist: the wet-chemical syn-
thesis and the condensation of nanoparticles in the gas phase. In this section, the
different methods are described and different possibilities to achieve FePt nanopar-
ticles in the chemically ordered state with an L10crystal symmetry are discussed.
These techniques need to be distinguished from techniques based on the annealing
or deposition of thin films which usually result in flat, pancake-like particles.12, 30
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1118 C. Antoniak & M. Farle
Ar/He
Nucleation chamber Sintering oven
Deposition
chamber
Pump
Fig. 5. Schematic drawing of the experimental setup for condensation of nanoparticles in the gas
phase.31
3.1. Gas phase condensation
FePt nanoparticles can be prepared by an inert gas condensation method based on
a DC magnetron sputtering process from alloy targets in a continuous gas flow of
helium and argon.31, 32 The experimental setup for preparing nanoparticles from the
gas phase is composed of three parts: a nucleation chamber, a sintering oven and a
deposition chamber as schematically shown in Fig. 5. After nucleation and particle
growth in the nucleation chamber with liquid nitrogen cooled walls, the particles
pass the sintering oven and can be in-flight annealed before deposited onto a sub-
strate. In order to obtain the chemically ordered L10phase in the nanoparticles,
annealing has to take place at very high temperatures around 1000-1300 K, since
the residence time of the particles in the hot zone is very short (about 1s or less
depending on the gas flow rate).
With this method it is possible to prepare chemically disordered FexPt1xpar-
ticles with diameters in the range of 3 nm < d < 20 nm that are single crystalline
or multiple twinned with an icosahedral shape. Size and morphology can be tuned
by changing the inert gas pressure and the sintering temperature.31 It was found,
that the icosahedral particles are thermally stable and can hardly be transformed
into the L10phase. This indicates an inadequate volume diffusion in the icosahedral
particles probably due to a lack of a sufficient amount of vacancies.
A method to destabilise the icosahedral shape and to promote the formation of
single-crystalline fcc FePt nanoparticles is the introduction of oxygen during the
particle preparation.33 However, the formation of the L10phase was not observed
indicating that the volume diffusion is still inadequate in this method. Diffusion
can be boosted by He-ion irradiation that may enhance the mobility of the Fe and
Pt atoms as known in the case of the irradiation of sputter- grown FePt films.34, 35
These results were also supported by Monte Carlo simulations.36 A further way
to enhance the diffusion in FePt is the introduction of nitrogen during the sputter
process and afterwards allow it to effuse by annealing. This method has been ap-
plied to drive the formation of the L10phase in FePt thin films37, 38 and adopted
successfully to the gas phase preparation of FePt nanoparticles.39
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1119
50nm
FFT
Fig. 6. Self-assembled Fe0.56 Pt0.44 nanoparticles on a Si substrate. The Fourier transformed
makes the hexagonal arrangement clear.
3.2. Wet-chemical synthesis
The wet-chemical synthesis was performed following the method published by
S. Sun et al.11 by the reduction of platinum diacetylacetonate (Pt(acac)2,
CH3COCHCOCH3) and thermal decomposition of iron pentacarbonyl (Fe(CO)5)
in hexadecane-1,2-diol at about 300C. The chemical reactions were initiated in the
presence of the surfactants oleic acid and oleyl amine, providing a route to synthesise
nanoparticles of a chemically disordered FexPt1xalloy surrounded by the surfac-
tants. After cooling down to room temperature, the particles were precipitated by
adding ethanol and separated by centrifugation. After that procedure, the parti-
cles were dispersed in n-hexane with surfactants, precipitated out and centrifuged
again. This can be repeated several times, until a stable dispersion of monodisperse
nanoparticles in n-hexane is obtained. Different compositions of the FexPt1xalloy
can be obtained by changing the molar ratio of Fe(CO)5to Pt(acac)2.
The nanoparticles were brought onto a naturally oxidised Si substrate using the
spin coating technique. The shell of organic ligands prevent the agglomeration of
the nanoparticles and drive the formation of hexagonally self-assembled superlat-
tices. The quality of the hexagonal arrangement can be improved by an excess of
surfactants in the dispersion.40 Subsequent annealing of the nanoparticles has been
tried as a route to obtain nanoparticles in the L10state.11, 41 Due to the thermal
decomposition of the ligand shell during the annealing process and the enhanced
mobility of the nanoparticles at elevated temperatures, this procedure leads to sin-
tering of the nanoparticles especially for small diameters below 6 nm. Much effort
has been dedicated to prevent sintering using different methods, e.g.:
(i) The nanoparticles can be linked to the substrate by special molecules like (3-
aminopropyl)dimethylethoxysilane (C7H19NOSi, APDMES)45 or other amino-
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1120 C. Antoniak & M. Farle
functional silanes.42 The linker molecules are brought onto the naturally oxi-
dised Si substrate and form a very stable Si-O bond. In the second step, the sur-
factants surrounding the nanoparticles are exchanged with amino-functional
groups of the linker-molecule layer surfaces.43, 44 As reported by M. Mizuno
et al.,45 after that preparation a sample consisting of FePt nanoparticles with
diameters around 4.4 nm can be annealed at 800C for 30 min without signif-
icant sintering.
(ii) A straight-forward method that has been successfully applied is the coverage
of the nanoparticle monolayer e.g. by carbon46 or bring them into a sodium
chloride (NaCl)-matrix by solving milled NaCl in the hexane in which the
nanoparticles are dispersed. After all of the solvent is evaporated, the powder
can be annealed and the NaCl can be washed away subsequently. Li et al.
showed,47 that for a NaCl : FePt ratio of 400:1, 4 nm FePt nanoparticles were
annealed at 700C for 8 h without significant sintering of the particles.
3.3. Plasma cleaning of nanoparticles
Due to their large surface-to-volume ratio, nanoparticles may exhibit a large fraction
of oxides. Even the organic ligand shell surrounding wet-chemically synthesised
nanoparticles, cannot protect them from oxidation48, 49 as indicated in Fig. 7(c) by
the doublet structure at both the Fe L3and L2absorption edges typical for Fe
oxides.50 The type of Fe oxide in the case of FePt nanoparticles is a mixture of
α-Fe2O3(hematite) and Fe3O4(magnetite).49 In the case of Fe nanoparticles, a
mixture of oxides was obtained as well.51
0.28 0.29 0.30 0.31
0.2
0.4
0.6
0.8
1.0
1.2
0
methan
ethan
ð*
Ry
ó*(C-C)
ó*(C=C)
(a)
(b)
photon energy, [keV]E
absorption [arb. units]
0.70 0.72 0.73
photon energy, [keV]E
0.71
0
0
2
4
6
8
10
absorption [arb. units]
2
4
6
(c)
(d)
Fig. 7. XANES at the carbon K edge (a)/(b) and at the iron L3,2edges (c)/(d) of FePt nanoparti-
cles in the as-prepared state (a)/(c) and after the hydrogen plasma treatment (b)/(d) that reduces
the oxides and removes the ligand shell as schematically drawn. For comparison two spectra of
alcanes57, 58 are shown.
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1121
One method to reduce the oxides at the surface of nanoparticles is the in situ
hydrogen plasma cleaning by means of a rf-plasma.52, 53 For wet-chemically synthe-
sised FePt nanoparticles with a mean diameter of 6.3 nm, after 15 min exposure
to a hydrogen plasma at a pressure of 5 Pa at room temperature, not only the
Fe oxides are reduced completely,54 but also the organic ligands are removed. This
was checked by the X-ray absorption near-edge spectroscopy (XANES) at the Fe
L3,2edges and the carbon K edge as shown in Fig. 7. The XANES spectrum at
the carbon K edge was normalised to a reference spectrum of a cleaned Si wafer. It
shows several absorption peaks55, 56 corresponding to electron excitations into π*
antibonding states at 285.1 eV, so-called Rydberg states at 288.1 eV (3s, 3d not
resolved energetically), σ* (C C) antibonding states of carbon-carbon single bonds
at 292.9 eV and σ(C C) antibonding states of carbon-carbon double bonds at
302 eV. After the sample was exposed for 15 min to the hydrogen plasma, no ab-
sorption peaks were observed indicating the complete removal of the ligands as also
known from reactive oxygen plasma.59 In addition, at the Fe sites pure metallic
spectra were obtained. Therefore, the in situ plasma cleaning procedure gives the
possibility to determine the properties of pure metallic nanoparticles without any
surface modifications.
4. Crystal Structure and Alloying
The determination of the lattice constant of 4.4 nm FePt nanoparticles by means
of X-ray diffraction (XRD) shows an enhanced value of aN P = (0.388 ±0.002) nm
with respect to the lattice constant in 50 nm thick FePt films of the same compo-
sition with a= (0.385 ±0.002) nm.60 Larger particles with a mean diameter of 6.3
nm exhibit the same lattice constant as the corresponding bulk material. HR TEM
images indicate, that there is a significant lattice expansion at the surface of icosa-
hedral FePt nanoparticles prepared by gas phase condensation, whereas the inner
part of the particle exhibits a contracted lattice constant compared to the bulk
alloy.61 Figure 8 shows the spacing of the lattice planes along the h111idirections
as a function of the distance from the particles’ surface for an icosahedron. In order
to exclude a lattice expansion that is only caused by the oxidation at the surface,
the lattice constant was checked by the analysis of the EXAFS oscillations at the
Pt L3edge of wet-chemically synthesised nanoparticles in the as-prepared state,
i.e. with Fe oxides at the surface, and after the reduction of the oxides by an in
situ hydrogen plasma treatment. In both cases, the lattice constant was enhanced,
aNP = (0.387 ±0.004) nm, in agreement with the value obtained by XRD, HR
TEM and electron diffraction (ED).
The EXAFS analysis was done by a simulation of the experimental data using
the Software Artemis63 based on the algorithms of the programmes FEFFIT64, 65
and FEFF.66 The simulations that fit best to the experimental data of the chemi-
cally disordered Fe0.56Pt0.44 nanoparticles and the reference spectra of a polished
Fe0.56Pt0.44 bulk sample are shown in Fig. 9. For the simulations, backscattering
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1122 C. Antoniak & M. Farle
0246
0.21
0.22
0.23
0.24
0.25
lattice parameter , [nm]d111
shell, n
(surface)
1st measurement
2nd measurement
3rd measurement
2nm
Fe Pt bulk
52 48
Fig. 8. HR TEM image of an icosahedral FePt nanoparticles with a fcc structure (left) and the
analysed spacing of the (111) lattice planes.62
-0.2
0
0.2 0.3
0.4
FT[ ( )] [arb. units]k k÷
20 40 60 80 100 120 140
wave number, [1/nm]k
0
k k÷( ) [arb. units]
00.2 0.4 0.6 0.8 1.0
radial distance, [nm]R
0
0.1
0.2
-0.2
0.2
0
0.1
0.2
bulk
bulk
nanoparticles
nanoparticles
Fig. 9. EXAFS oscillations as a function of wave number (left) and Fourier transformed (right)
of Fe0.56Pt0.44 bulk material and nanoparticles. Open symbols refer to experimental data, solid
lines to 1st shell simulations.
from the nearest neighbour atoms of the absorber Pt atom only was included. Since
the sample is chemically disordered, the intensity was calculated assuming that the
first shell of neighbouring atoms contains either Pt atoms or Fe atoms only. The
relative weight of these two contributions to the total absorption spectrum was a
fit parameter reflecting the composition of the nearest neighbours of the Pt ab-
sorber atoms. In the case of the bulk material, this composition was equal to the
averaged composition of the whole sample measured by the energy-dispersive X-ray
spectroscopy (EDS) within the error bars. For the nanoparticles, a strong deviation
from the averaged composition was obtained, indicating that Pt is in a Pt rich
environment and as a consequence, Fe is in an Fe rich environment. The values of
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1123
Table 1. Lattice constant and composition of FePt nanoparticles and the corresponding
bulk alloy. The composition obtained by EXAFS describes only the composition around the
Pt absorber atoms (1st shell).
Bulk material Nanoparticles
Lattice constant (nm) 0.38167 0.388 ±0.002 (XRD)
0.381 ±0.003 (Pt EXAFS) 0.387 ±0.004 (Pt EXAFS)
0.389 ±0.006 (HR TEM, ED)
Fe content (at%) 56 ±2 (EDS) 56 ±5 (EDS)
54 ±4 (EXAFS) 38 ±8 (Pt EXAFS)
the lattice constant and the composition of the Pt nearest environment obtained
by EXAFS are listed in Table 1 and compared to the values obtained by different
experimental methods.
5. Magnetism
In this section, we focus on the element-specific magnetic moments in FePt nanopar-
ticles and their dependence on size and crystal structure, that have been determined
by means of the XANES and its associated X-ray magnetic circular dichroism
(XMCD).68 The measurements at the Fe L3,2absorption edges in the soft X-ray
regime were performed at the PM3 bending magnet beamline of the BESSY II syn-
chrotron radiation source in Berlin, Germany, in magnetic fields of up to ±2.8 T
in the total electron yield (TEY) mode. Absorption spectra at the Pt L3,2edges in
the hard X-ray regime were taken at the ID12 undulator beamline of the European
Synchrotron Radiation Facility (ESRF) in Grenoble, France, in fields of up to ±0.6
T in fluorescence yield (FY) mode. After each scan, either the magnetic field or the
X-ray polarisation was flipped.
To separate the transitions into the 3d states of the Fe atoms and into the 5d final
states of the Pt atoms from transitions into higher levels (or into the continuum), a
two-step like function was subtracted in the case of the XANES at the Fe L3,2edges.
Since the absorption at the Pt L3,2edges is not well-pronounced, this procedure
would lead to a large error. Therefore, reference spectra of Au are shifted in energy
and subtracted instead69 after stretching the energy scale to account for the different
lattice constant in Au and FePt. From the XANES and XMCD spectra, the effective
spin magnetic moment µeff
Sand the orbital magnetic moment µlcan be determined
according to the sum rules,70, 71 that were experimentally confirmed for the 3d
transition metals.72 For the numbers of unoccupied final states 3.41 at the Fe sites
and 1.74 at the Pt sites were used obtained from band structure calculations.73
Note, that the effective spin magnetic moment µeff
S=µs+ 7µtconsists of the
spin magnetic moment and a magnetic dipole moment accounting for an asphericity
of the spin density distribution. In systems with a weak SOC, the contribution of
µtcan be separated by angular dependent measurements of the XMCD.74 Since
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1124 C. Antoniak & M. Farle
bulk 6nm NP 4.5nm NP 3.4nm NP
effective spin magn. mom., [ ]μSμB
eff
0.40
0.45
0.50
2.0
2.5
3.0
Fe
Pt
Fig. 10. Effective spin magnetic moments of Fe and Pt in FePt bulk material and nanoparticles
with different sizes.
the SOC in FePt is not weak, even in an ensemble of nanoparticles with randomly
distributed crystallographic axes, µtmay not average out.75
The absorption spectra recorded in the TEY mode are influenced by saturation
effects76 that lead to an underestimation of the magnetic moments also in the case
of nanoparticles.77 Therefore, the effective spin magnetic moments were corrected
by 2% in 6 nm FePt nanoparticles and the orbital magnetic moments by about
20%.
Since the XMCD is proportional to the magnetisation, element-specific magneti-
sation curves were recorded at the Fe L3edges by measuring the field-dependent
XMCD signal78 at the energy where the dichroism was found to be maximum
(707.8 eV) and normalisation to the absorption in the pre-edge region at 700 eV.
5.1. Size-dependence of element-specific magnetic moments
The size-dependence of the magnetic moments was analysed for a composition
around 50 at.% Fe content. The values are listed in Table 2 and the effective spin
magnetic moments at the Fe sites are shown in Fig. 10. Compared to the corre-
sponding bulk material with µeff
S= (2.92 ±0.29)µBand µl= (0.083 ±0.012)µBat
the Fe sites and µeff
S= (0.47 ±0.02)µBand µl= (0.045 ±0.006)µBat the Pt sites,
the effective spin magnetic moment is reduced by about 20% at the Fe sites and
by 13% at the Pt sites of FePt nanoparticles with a mean diameter of 6.3 nm. This
reduction can be explained by the inhomogeneous alloying within the nanoparticles
as indicated by the analysis of EXAFS oscillations (Sec. 4). An Fe rich environment
of Fe atoms leads to a reduced spin magnetic moment. In the extreme case of an fcc
Fe sample, the average total magnetic moment vanishes due to an antiferromagnetic
spin structure. Since the spin magnetic moments at the Pt sites are induced by the
Fe spin magnetic moments, Pt atoms in a Pt rich environment may exhibit smaller
moments as well. In the literature, reductions of the spin magnetic moment are also
discussed in terms of spin canting effects53 due to a large surface anisotropy with
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1125
Table 2. Effective spin magnetic moment and orbital magnetic moment at the
Fe sites determined from the analysis of the XMCD for FePt nanoparticles (NP)
of different sizes.
Composition Diameter (nm) µeff
S(µB)µl(µB)µeff
Sl(Fe)
Fe0.50Pt0.50 6.3 2.28 ±0.25 0.048 ±0.010 2.1%
Fe0.56Pt0.44 4.4 2.13 ±0.21 0.062 ±0.014 2.9%
Fe0.48Pt0.52 3.4 2.01 ±0.16 0.068 ±0.015 3.4%
respect to the exchange coupling79, 80 or negative contributions of µtthat increases
at the surface as in the case of Fe clusters.81 These effects cannot be excluded but
are expected to be too small to explain the strong reduction of µswith respect
to the bulk material. However, they might be the reason for the further decrease
of µeff
Sat the Fe sites with decreasing particle size to (2.01 ±0.16)µBfor particles
with a mean diameter of 3.4 nm. In addition, a small change in the number of
unoccupied final states may be present, that was assumed to be constant. The ratio
of orbital-to-effective-spin magnetic moment increases with deceasing particle size
as qualitatively expected due to the break of symmetry at the surface. In the case
of the Fe atoms, the increase is from 2.1% for 6.3 nm particles to 3.4% for 3.4 nm
particles. These ratios are independent from the number of unoccupied final states
and therefore, this change is significant. As discussed in the following section, the
lowering of the crystal symmetry due to chemical order will increase µeff
Sl(Fe)
even more to about 9%.
5.2. Influence of chemical order on magnetic properties
In order to achieve the chemically ordered phase wet-chemically synthesised and
plasma cleaned FePt nanoparticles with a mean diameter of 6.3 nm were annealed
in situ at 600C in a hydrogen atmosphere of 5 Pa for 30 min. SEM images show
that after this procedure, more than 80% of the sample still consist of well-separated
nanoparticles. The mean diameter remains largely unchanged at 6.43 nm, only the
variance of the size distribution increased from σ= 0.14 to σ= 0.25. For the
ensemble of FePt nanoparticles, the magnetic moments at both the Fe and Pt
sites were determined and the field-dependent magnetisation was recorded at the
Fe L3edge at different temperatures before and after annealing. The results54 are
compared to the magnetic moments at the Fe sites of gas phase synthesised particles
which were in-flight annealed. The element-specific magnetisation curves are shown
in Fig. 11 for the external magnetic field perpendicular to the sample plane (θ= 0)
and for θ= 75. Before annealing, the coercive field is µ0Hc= (36 ±5) mT and
due to the magnetic dipolar interactions between the particles, the hard axis of the
magnetisation is perpendicular to the sample plane. After annealing, the coercive
field exhibits a slight dependence on the polar angle and is (292 ±8) mT for θ= 0
and (228 ±8) mT for θ= 75. These values are in agreement to coercive fields of
similar particle ensembles.32 At 305 K a coercive field of (35 ±5) mT was obtained
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1126 C. Antoniak & M. Farle
magnetisation, /M MS
0
0.5
1.0
-0.5
-1.0
magnetisation, /M MS
0
0.5
1.0
-0.5
-1.0
-1.0 -0.5 1.00.50
magnetic field, [T]µ H
0
-3 -2 320
magnetic field, [T]µ H
0
1
-1
= 0°
= 75°
è
è
= 0°
= 75°
è
è
A1 L10
Fig. 11. Field-dependent magnetisation measured at the Fe L3edge at T= 15 K before (left)
and after annealing (right).
050 100 150 200 250 300 350
temperature, [K]T
0.0
0.1
0.2
0.3
0.4
coercive field, [T]µ H
0C
µ H
0C(0)
TB
mean
510 15 20 25 30
T K
2/3 2/3
[ ]
0.0
0.1
0.2
0.3
coercive field, [T]µ H
0C
Fig. 12. Temperature-dependence of the measured coercive field (circles) and calculated accord-
ing to Eq. (10).
in agreement to the coercive field of (38 ±7) mT of a gas phase prepared sample
consisting of in-flight annealed particles with a mean diameter around 6 nm and a
fraction of 70% of the particles in the L10phase.82
The shape of the magnetisation curve is largely independent of the angle be-
tween the sample plane and the external magnetic field, indicating that the magnetic
dipolar coupling of the particles is negligible with respect to the anisotropy of the
single particles. The ratio of remanence-to-saturation magnetisation is 0.5 as pre-
dicted in the model of Stoner and Wohlfarth for non-interacting particles with
uniaxial anisotropy and randomly oriented easy axes. In this case, the tempera-
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1127
×5
×5
×20
13.28 13.32
11.60
11.56
0.74
0.72
0.70
photon energy, [keV]E
-4
-2
0
2
4
6
×5
×5
×20
-6
-3
0
3
6
9
XMCD integral [arbitrary units]
-4
-2
0
2
4
6
intensity [arbitrary units]
dE(XMCD)
ò
XANES
XMCD
-6
-3
0
3
6
9
A1
L10
Fig. 13. XANES and XMCD at the Fe and Pt L3,2edges of wet-chemically synthesised FePt
nanoparticles before (top) and after annealing (bottom). Grey lines show the integrated XMCD
signal, dashed lines represent the spectra of in-flight annealed FePt particles from the gas phase.
ture dependence of the coercive field follows Sharrock’s law and can be calculated
according to83
Hc(T)Hc(0) "125kBT
Keff V1/m#(10)
where m= 3/2 for randomly oriented easy axes of the magnetisation. The experi-
mentally obtained values of the coercive field at different temperatures well below
the mean blocking temperature, can be fitted to Eq. (10) using µ0HC0.36 T
and Keff 4.7×105J/m3that is about one order of magnitude smaller than
the anisotropy of bulk FePt in the L10phase. This may be due to a fraction of
superparamagnetic fcc particles in the sample and/or a lower degree of chemical
ordering.
From measurements of the XANES and the XMCD as shown in Fig. 13, the
element-specific magnetic moments were calculated. The chemical ordering and
the induced tetragonal distortion of the otherwise cubic lattice also influences the
element-specific magnetic moments. The values are listed in Table 3 and are a sen-
sitive monitor to structural changes. The effective spin magnetic moment at the Fe
sites increases slightly, that may confirm the inhomogeneous alloying in the chem-
ically disordered A1 phase, i.e. the formation of Fe rich and Pt rich parts within
the single particles. After annealing these inhomogeneities should vanish due to
the chemical ordering and µeff
Sincreases. Note, that the sample might be magnet-
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1128 C. Antoniak & M. Farle
Table 3. Element-specific magnetic moments of FePt bulk material and nanoparticles (NP) in
the A1 state, wet-chemically synthesised nanoparticles in the (partial) L10state and nanopar-
ticles in the L10state from the gas phase (last row).
Fe Pt
µeff
s(µB)µl(µB)µeff
s(µB)µl(µB)µleff
s(%)
Bulk (A1) 2.92 ±0.29 0.083 ±0.012 0.47 ±0.02 0.045 ±0.006 3.6±1
NP (A1) 2.28 ±0.25 0.048 ±0.010 0.41 ±0.02 0.054 ±0.006 3.8±1
NP (L10) 2.38 ±0.26 0.204 ±0.020 0.41 ±0.04 0.042 ±0.008 8.8±1
NP (L10)82 2.21 ±0.20 0.194 ±0.020 —
ically not fully saturated in a magnetic field of 2.8 T as the magnetisation curve
(Fig. 11) suggests. This may be the reason, why the increase of the Fe magnetic
moment is rather small and at the Pt sites, no change is observed for µeff
Sand µl
slightly decreases. The most significant change after the annealing procedure can
be observed at µlof the Fe sites: it increases by almost a factor of 4 to a value of
(0.204 ±0.020)µBand indicates the deviation from the cubic symmetry. In the case
of the sample consisting of in-flight annealed nanoparticles with a fraction of 70%
in the L10state, the same value of µlwithin the error bars is obtained.
6. Summary
We presented the dependence of the element-specific magnetic moments on size and
structure of hydrogen plasma cleaned FePt nanoparticles. The reduction of the ef-
fective spin magnetic moment µeff
S2.28µBat the Fe sites of 6.3 nm FePt particles
with respect to the corresponding bulk material (µeff
s2.92µB) can be explained
by the imperfect alloying suggested by the analyses of the EXAFS oscillations at
the Pt L3edges. For smaller particle size µeff
Sdecreases to µeff
S2.01µBfor 3.4
nm particles and the ratio of orbital to effective spin magnetic moment µleff
sin-
creases from 2.1% to 3.4%. The latter is qualitatively expected due to the break of
symmetry at the surface.
A lowering of the crystal symmetry from the A1 to (partial) L10structure yields
an increase of µleff
sto about 9% in the case of 6.3 nm particles due to a fourfold
enhanced µlat the Fe sites. The transformation from the state with A1 symmetry
to the (partially) L10ordered one is accompanied by an increase of the coercive
field by a factor of 8 to (292±8) mT at T= 15 K and θ= 0. From the temperature
dependence of the coercive field the effective anisotropy of the L10particles could
be estimated: Keff 4.7×105J/m3. Although this value is one order smaller than
the anisotropy of the bulk material, there is still a coercive field of µ0Hc35 mT at
T= 305 K. These results on wet-chemically synthesised nanoparticles which were
annealed after deposition onto a substrate correspond very well to the results on in-
flight annealed nanoparticles of the same size prepared by gas phase condensation.
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1129
Acknowledgments
For sample preparation we thank S. Sun (Brown U.), V. Salgueiri˜no-Maceira (U. de
Santiago de Compostela) and O. Dmitrieva, who is also acknowledged for structural
characterisation together with A. Schlachter and D. Sudfeld (U. Duisburg-Essen).
For help during the synchrotron beamtimes we thank K. Fauth (MPI for Metals
Research), U. Wiedwald, H.-G. Boyen (U. Ulm), M. Spasova, M. Acet, A. Trunova,
N. Friedenberger, S. Stienen (U. Duisburg-Essen) and the BESSY II staff, especially
Th. Kachel and H. Pfau, as well as F. Wilhelm, A. Rogalev, P. Voisin, and S. Feite
from the ESRF.
This work was financially supported by the DFG (SFB445), the BMBF (05
ES3XBA/5), the ESRF, and the EU (MRTN-CT-2004-0055667, “SyntOrbMag”).
References
1. X. Batlle and A. Labarta, J. Phys. D: Appl. Phys. 35 (2002) R15.
2. U. Wiedwald, J. Lindner, M. Spasova, Z. Frait and M. Farle, Phase Transitions 78
(2005) 85.
3. U. Wiedwald, M. Spasova, E. L. Salabas, M. Ulmeanu, M. Farle, Z. Frait, A. Fraile
Rodriguez, D. Arvanitis, N. S. Sobal, M. Hilgendorff and M. Giersig, Phys. Rev. B 68
(2003) 064424.
4. M. Spasova and M. Farle, Low-Dimensional Systems: Theory, Preparation, and some
Applications, eds. L. Marz´an and M. Giersig, NATO Science Series II, Vol. 91
(Springer, 2003) p. 173.
5. O. Margeat, M. Tran, M. Spasova and M. Farle, Phys. Rev. B 75 (2007) 134410.
6. B. Predel, Landolt-B¨ornstein: Numerical Data and Functional Relationships in
Science and Technology, New Series IV/5e, ed. O. Madelung (Springer, Berlin, 1995)
p. 222.
7. B. Rellinghaus, J. K¨astner, T. Schneider, E. F. Wassermann and P. Mohn, Phys. Rev.
B51 (1995) 2983.
8. S. H. Whang, Q. Feng and Y.-Q. Gao, Acta Mater. 46 (1998) 6485.
9. M. R. Visokay and R. Sinclair, Appl. Phys. Lett. 66 (1995) 1692.
10. J.-U. Thiele, L. Folks, M. F. Toney and D. K. Weller, J. Appl. Phys. 84 (1998) 5686.
11. S. Sun, C. B. Murray, D. Weller, L. Folks and A. Moser, Science 287 (2000) 1989.
12. T. Shima, K. Takanashi, Y. K. Takahashi and K. Hono, Appl. Phys. Lett. 85 (2004)
2571.
13. S. Sun and C. B. Murray, J. Appl. Phys. 85 (1999) 4325.
14. J. L. Dorman, D. Fiorani and E. Tronc, Adv. Chem. Phys. 98 (1997) 283.
15. G. van der Laan, J. Phys.: Cond. Matter 10 (1998) 3239.
16. B. ´
Ujfalussy, L. Szunyogh, P. Bruno and P. Weinberger, Phys. Rev. Lett. 77 (1996)
1805.
17. L. Szunyogh, B. ´
Ujfalussy and P. Weinberger, Phys. Rev. B 51 (1995) 9552.
18. R. Wu, L. Chen and A. J. Freeman, J. Magn. Magn. Mat. 170 (1997) 103.
19. M. Spasova, U. Wiedwald, R. Ramchal, M. Farle, M. Hilgendorff and M. Giersig,
J. Magn. Magn. Mater. 240 (2002) 40.
20. L. N´eel, J. Phys. Rad. 15 (1954) 225.
21. P. Bruno, Phys. Rev. B 39 (1989) 865.
22. M. Jamet, W. Wernsdorfer, C. Thirion, V. Dupuis, P. M´elinon, A. erez and
D. Mailly, Phys. Rev. B 69 (2004) 024401.
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
1130 C. Antoniak & M. Farle
23. C. Herring, Phys. Rev. 82 (1951) 87, and references therein.
24. L. N´eel, Ann. Geophys. (CNRS) 5(1949) 99.
25. W. F. Brown, J. Appl. Phys. 30 (1959) 130S.
26. M. Respaud, J. M. Broto, H. Rakoto, A. R. Fert, L. Thomas, B. Barbara, M. Verelst,
E. Snoeck, P. Lecante, A. Mosset, J. Osuna, T. Ould Ely, C. Amiens and B. Chaudret,
Phys. Rev. B 57 (1998) 2925.
27. E. P. Wohlfarth, Phys. Lett. A 70 (1979) 489.
28. C. Antoniak, J. Lindner and M. Farle, Europhys. Lett. 70 (2005) 250.
29. K. D. Usadel, Phys. Rev. B 73 (2006) 212405.
30. G. Q. Li, H. Takahoshi, H. Ito, H. Saito, S. Ishio, T. Shima and K. Takanashi,
J. Appl. Phys. 94 (2003) 5672.
31. S. Stappert, B. Rellinghaus, M. Acet and E. F. Wassermann, J. Cryst. Growth 252
(2003) 440.
32. J.-M. Qiu, J. H. Judy, D. Weller and J.-P. Wang, J. Appl. Phys. 97 (2005) 10J319.
33. S. Stappert, B. Rellinghaus, M. Acet and E. F. Wassermann, Eur. Phys. J. D 24
(2003) 351.
34. D. Ravelosona, C. Chappert, V. Mathet and H. Bernas, Appl. Phys. Lett. 76 (2000)
236.
35. C.-H. Lai, C.-H. Yang and C. C. Chiang, Appl. Phys. Lett. 83 (2003) 4550.
36. H. Bernas, J.-Ph. Attane, K.-H. Heinig, D. Halley, D. Ravelosona, A. Marty, P. Auric,
C. Chappert and Y. Samson, Phys. Rev. Lett. 91 (2003) 77203.
37. H. Y. Wang, W. H. Mao, X. K. Ma, H. Y. Zhang, Y. B. Chen, Y. J. He and E. Y.
Jiang, J. Appl. Phys. 95 (2004) 2564.
38. H. H. Hsiao, R. N. Panda, J. C. Shih, and T. S. Chin, J. Appl. Phys. 91 (2002) 3145.
39. O. Dmitrieva, M. Acet, G. Dumpich, J. K¨astner, C. Antoniak, M. Farle, and K. Fauth,
J. Phys. D: Appl. Phys. 39 (2006) 4741.
40. N. Shukla, J. Ahner and D. Weller, J. Magn. Magn. Mater. 272 (2004) e1349.
41. Z. Jia, S. Kang, S. Shi, D. E. Nikles and J. W. Harrell, J. Appl. Phys. 97 (2005)
10J310.
42. A. C. C. Yu, M. Mizuno, Y. Sasaki, M. Inoue, H. Kondo, I. Ohta, D. Djayaprawira
and M. Takahashi, Appl. Phys. Lett. 82 (2003) 4352.
43. V. L. Colvin, A. N. Goldstein and A. P. Alivisatos, J. Am. Chem. Soc. 114 (1992)
5221.
44. S. Sun, S. Anders, H. F. Hamann, J.-U. Thiele, J. E. E. Baglin, T. Thomson, E. E.
Fullerton, C. B. Murray and B. D. Terris, J. Am. Chem. Soc. 124 (2002) 2884.
45. M. Mizuno, Y. Sasaki, A. C. C. Yu and M. Inoue, Langmuir 20 (2004) 11305.
46. M.-P. Chen, H. Nishio, Y. Kitamoto and H. Yamamoto, J. Appl. Phys. 97 (2005)
10J321.
47. D. Li, N. Poudyal, V. Nandwana, Z. Jin, K. Elkins and J. P. Liu, J. Appl. Phys. 99
(2006) 08E911.
48. S. Anders, M. F. Toney, T. Thomson, J.-U. Thiele, B. D. Terris, S. Sun and C. B.
Murray, J. Appl. Phys. 93 (2003) 7343.
49. C. Liu, T. J. Klemmer, N. Shukla, X. Wu, D. Weller, M. Tanase and D. Laughlin,
J. Magn. Magn. Mater. 266 (2003) 96.
50. T. J. Regan, H. Ohldag, C. Stamm, F. Nolting, J. L¨uning, J. St¨ohr and R. L. White,
Phys. Rev. B 64 (2001) 214422.
51. K. Fauth, E. Goering, G. Sch¨utz and L. Theil Kuhn, J. Appl. Phys. 96 (2004) 399.
52. H.-G. Boyen, G. K¨astle, K. Z¨urn, T. Herzog, F. Weigl, P. Ziemann, O. Mayer, C.
Jerome, M. M¨oller, J. P. Spatz, M. G. Garnier and P. Oelhafen, Adv. Func. Mater.
13 (2003) 359.
Final Reading
August 7, 2007 15:42 WSPC/147-MPLB 01382
Magnetism at the Nanoscale: The Case of FePt 1131
53. H.-G. Boyen, K. Fauth, B. Stahl, P. Ziemann, G. K¨astle, F. Weigl, F. Banhart,
M. Hessler, G. Sch¨utz, N. S. Gajbhiye, J. Ellrich, H. Hahn, M. B¨uttner, M. G. Garnier
and P. Oelhafen, Adv. Func. Mater. 17 (2005) 574.
54. C. Antoniak, J. Lindner, M. Spasova, D. Sudfeld, M. Acet, M. Farle, K. Fauth,
U. Wiedwald, H.-G. Boyen, P. Ziemann, F. Wilhelm, A. Rogalev and S. Sun, Phys.
Rev. Lett. 97 (2006) 117201.
55. J. St¨ohr, F. Sette and A. L. Johnson, Phys. Rev. Lett. 53 (1984) 1684.
56. A. P. Hitchcock, S. Beaulieu, T. Steel, J. St¨ohr and F. Sette, J. Chem. Phys. 80
(1984) 3927.
57. G. R. Wight and C. E. Brion, J. Electron Relat. Phenom. 4(1974) 25.
58. A. P. Hitchcock and C. E. Brion, J. Electron Relat. Phenom. 10 (1977) 317.
59. U. Wiedwald, K. Fauth, M. Heßler, H.-G. Boyen, F. Weigl, M. Hilgendorff, M. Giersig,
G. Sch¨utz, P. Ziemann and M. Farle, Chem. Phys. Chem. 6(2005) 2522.
60. A. Trounova, Diploma Thesis, Universit¨at Duisburg-Essen (2004).
61. N. Friedenberger, Diploma Thesis, Universit¨at Duisburg-Essen (2007).
62. R. Wang, O. Dmitrieva, M. Farle, G. Dumpich, H. Poppa, R. Kilaas, and
C. Kisielowski, submitted.
63. B. Ravel and M. Newville, J. Synchr. Rad. 12 (2005) 537.
64. M. Newville, J. Synchr. Rad. 8(2001) 322.
65. A. L. Ankudinov, B. Ravel, J. J. Rehr and S. D. Conradson, Phys. Rev. B 58 (1998)
7565.
66. S. I. Zabinsky, J. J. Rehr, A. L. Ankudinov, R. C. Albers and M. J. Eller, Phys. Rev.
B52 (1995) 2995.
67. B. Predel, Landolt-B¨ornstein: Numerical data and functional relationships in science
and technology, New Series IV/5e, ed. O. Madelung (Springer, Berlin, 1995) p. 224.
68. G. Sch¨utz, W. Wagner, W. Wilhelm, P. Kienle, R. Zeller, R. Frahm and G. Materlik,
Phys. Rev. Lett. 58 (1987) 737.
69. W. Grange, M. Maret, J.-P. Kappler, J. Vogel, A. Fontaine, F. Petroff, G. Krill,
A. Rogalev, J. Goulon, M. Finazzi and N. B. Brookes, Phys. Rev. B 58 (1998) 6298.
70. B. T. Thole, P. Carra, F. Sette and G. van der Laan, Phys. Rev. Lett. 68 (1992) 1943.
71. P. Carra, B. T. Thole, M. Altarelli, and X. Wang, Phys. Rev. Lett. 70 (1993) 694.
72. C. T. Chen, Y. U. Idzerda, H.-J. Lin, N. V. Smith, G. Meigs, E. Chaban, G. H. Ho,
E. Pellegrin, and F. Sette, Phys. Rev. Lett. 75 (1995) 152.
73. H. Ebert et al., The Munich SPR-KKR-Package, version 3.6, http://olymp.cup.uni-
muenchen.de/ak/ebert/SPRKKR; H. Ebert, in Electronic Structure and Physical
Properties of Solids, ed. H. Dreyse´e, Lecture Notes in Physics 535 (Springer), p. 191.
74. J. St¨ohr and H. K¨onig, Phys. Rev. Lett. 75 (1995) 3748.
75. C. Ederer, M. Komelj and M. F¨ahnle, Phys. Rev. B 68 (2003) 052402.
76. R. Nakajima, J. St¨ohr and Y. U. Idzerda, Phys. Rev. B 59 (1999) 6421.
77. K. Fauth, Appl. Phys. Lett. 85 (2004) 3271.
78. E. Goering, A. Fuss, W. Weber, J. Will and G. Sch¨utz, J. Appl. Phys. 88 (2000) 5920.
79. Y. Labaye, O. Crisan, L. Berger, J. M. Greneche and J. M. D. Coey, J. Appl. Phys.
91 (2002) 8715.
80. D. A. Garanin and H. Kachkachi, Phys. Rev. Lett. 90 (2003) 065504.
81. K. W. Edmonds, C. Binns, S. H. Baker, S. C. Thornton, C. Norris, J. B. Goedkoop,
M. Finazzi and N. B. Brookes, Phys. Rev. B 60 (1999) 472.
82. O. Dmitrieva, M. Spasova, C. Antoniak, M. Acet, J. K¨astner, G. Dumpich, M. Farle,
K. Fauth, U. Wiedwald, H.-G. Boyen, and P. Ziemann, submitted.
83. M. P. Sharrock, J. Appl. Phys. 76 (1994) 6413.
... Moreover, it is argued that the magnetism of Pt atoms is purely induced and hence, contributes to the FM order [29][30][31][32]. This is in good agreement with the several experimental evidence, notably by neutron diffraction, X-ray magnetic circular dichroism (XMCD) measurements [33,34], and ground state DFT calculations [24,27]. ...
... The relative orientations of Fe and Pt magnetic moments have also been discussed. While XMCD measurements clearly favor parallel (FM) alignment [34], studies have reported neutron diffraction on Fe 72 Pt 28 [35] and FePt [36] suggest antiparallel (ferrimagnetic, FIM) ordering of the Fe and Pt magnetic moments to predict the magnetic properties of FePt nanoparticles as a function of their size [19]. In contrast, another neutron diffraction study [33] reported a better fit of the experimental data when a parallel orientation of the Fe and Pt moments was assumed. ...
... In contrast, another neutron diffraction study [33] reported a better fit of the experimental data when a parallel orientation of the Fe and Pt moments was assumed. First-principles calculations of FePt within an FM phase reported a parallel orientation of the Fe and Pt magnetic moments [23][24][25]30,33], which was in agreement with the XMCD results [34] and the latter neutron diffraction study [33]. The influence of size effects on the magnetism of nanostructured FePt alloys has also been studied [18,19,26], and demonstrated important consequences for their applicability in high-density magnetic recording. ...
Article
Full-text available
In this study, we applied electron energy-loss magnetic chiral dichroism (EMCD), an electron counterpart of X-ray magnetic circular dichroism (XMCD), to a network nanostructured FePt L10 ordered alloy film to examine the relative orientation of magnetic moments between neighboring Fe and Pt atoms using the Fe-M2,3, Pt-O2,3, and Pt-N6,7 semi-core excitation spectra with transmission electron microscopy and electron energy-loss spectroscopy. EMCD signals were successfully extracted from a large number of spectra using a dedicated data analysis procedure to obtain sufficient noise statistics. Results showed that the relative sign relation of the EMCD signals between the Fe and Pt absorption edges was consistent with that of the theoretical dielectric tensor while assuming that parallel magnetic moments exist between neighboring Fe and Pt. We believe the results of this study can be applied to alloys with different nanostructures to determine whether the spin configuration depends on the size and geometry of the nanostructures.
... A simple model of non-interacting single domain particles [1] has been used for the fit of experimental M zfc (T)/M s dependences shown in Fig. 6. The population of MNPs (given by a volume distribution f(V)) is sharply divided into two groups at each temperature, depending on their particular size-the fraction in an ideal superparamagnetic state that corresponds to MNPs below a certain critical volume and those, above such limit, whose magnetic moment remains blocked [23]: ...
Article
Full-text available
Two sets of core/shell magnetic nanoparticles, CoFe2O4/Fe3O4 and Fe3O4/CoFe2O4, with a fixed diameter of the core (~ 4.1 and ~ 6.3 nm for the former and latter sets, respectively) and thickness of shells up to 2.5 nm were synthesized from metal chlorides in a diethylene glycol solution. The nanoparticles were characterized by X-ray diffraction, transmission electron microscopy, and magnetic measurements. The analysis of the results of magnetic measurements shows that coating of magnetic nanoparticles with the shells results in two simultaneous effects: first, it modifies the parameters of the core-shell interface, and second, it makes the particles acquire combined features of the core and the shell. The first effect becomes especially prominent when the parameters of core and shell strongly differ from each other. The results obtained are useful for optimizing and tailoring the parameters of core/shell spinel ferrite magnetic nanoparticles for their use in various technological and biomedical applications.
... The published studies on two-dimensional arrays of ferromagnetic (FM) nanoparticles usually report spherical particle form [3,[12][13][14]. Due to limitations of the experiment, magnetic characterization provides a collective signature for particle cluster that may vary with particle size [5,13], or crystallographic directions. These circumstances complicate the characterization of individual nanoparticles, providing instead ensemble-averaged char-Correspondence to: paul.horley@cimav.edu.mx ...
Article
Full-text available
We present a detailed theoretical study of the role of long-range dipole-dipole interactions on the angular dependence of ferromagnetic resonance spectra in a two-dimensional array of nanocubes. Variations of polar (ϕ) and azimuthal (θ) angles are studied numerically and analytically to illustrate the effect of the magnetocrystalline properties and the dipole-dipole interactions, forming complex resonance bands. In addition, we show that when the static magnetic field lies in the arrays' plane under the angle of 129° with the edge of the array or when its tilted around 15° to the plane's normal, the spectra of absorption transform into a plateau spanning from 0.1 T to 0.4 T, which is prominent enough for experimental observation.
... These ligands act as spacers between the nanoparticles and avoid the formation of agglomerates. They can be removed by a soft hydrogen plasma treatment which reduces also iron oxides that may be present at the surface [151,152]. Subsequently, the chemically ordered state of the FePt alloy is achieved by annealing the sample. As discussed above in section 3.2, special care has to be taken to exclude sintering of the nanoparticles at elevated temperatures. ...
Article
Full-text available
X-ray absorption spectroscopy facilitated by state-of-the-art synchrotron radiation technology is presented as a powerful tool to study nanoscale systems, in particular revealing their static element-specific magnetic and electronic properties on a microscopic level. A survey is given on the properties of nanoparticles, nanocomposites and thin films covering a broad range of possible applications. It ranges from the ageing effects of iron oxide nanoparticles in dispersion for biomedical applications to the characterisation on a microscopic level of nanoscale systems for data storage devices. In this respect, new concepts for electrically addressable magnetic data storage devices are highlighted by characterising the coupling in a BaTiO3/CoFe2O4 nanocomposite as prototypical model system. But classical magnetically addressable devices are also discussed on the basis of tailoring the magnetic properties of self-assembled ensembles of FePt nanoparticles for data storage and the high-moment material Fe/Cr/Gd for write heads. For the latter cases, the importance is emphasised of combining experimental approaches in x-ray absorption spectroscopy with density functional theory to gain a more fundamental understanding.
Article
In this work, the phase stability and properties of FeRh1-xPtx alloys (x = 0.125, 0.25, 0.375, 0.5, 0.625, 0.75, 0.875, 1) are studied within the framework of the density functional theory implemented in the VASP software package. It is found that the antiferromagnetic spin configurations AFM-II and AFM-III are energetically favorable for FeRh1-xPtx with a Pt content in the range of 0 < x ≤ 0.625 and 0.625 < x < 1, correspondingly. While, binary FePt alloy is ordered ferromagnetically. It is shown that compounds with 0 < x ≤ 0.625 are found to be stable with respect to decomposition into stable binary FeRh and FePt compounds at zero and finite temperatures. In the case of compositions with x = 0.75 and 0.875, their stability against segregation is predicted at finite temperatures. An increase in temperature enhances the stabilization of FeRh1-xPtx with respect to decomposition into two-component mixture.
Article
Magnetic properties of the sets of Fe3O4(core)/CoFe2O4(shell) composite nanoparticles with a core diameter of about 6.3 nm and various shell thicknesses (0, 1.0, and 2.5 nm), as well as the mixtures of Fe3O4 and CoFe2O4 nanoparticles taken in the ratios corresponding to the core/shell material contents in the former case, have been studied. The results of magnetic research showed that the coating of magnetic nanoparticles with a shell gives rise to the appearance of two simultaneous effects: the modification of the core/shell interface parameters and the parameter change in both the nanoparticle’s core and shell themselves. As a result, the core/shell particles acquire new characteristics that are inherent neither to Fe3O4 nor to CoFe2O4. The obtained results open the way to the optimization and adaptation of the parameters of the core/shell spinel-ferrite-based nanoparticles for their application in various technological and biomedical domains.
Article
Nanocomposites based on half-metallic FePt (L10) magnetic nanoparticles coated with the semiconducting conjugated polymer poly(3-hexylthiophene-2,5-diyl) (P3HT) show a significant reduction in the magnetic coercivity. This study adopts a physical approach based on chemical potential equalization at the interface. The underlying charge/spin transfer mechanism unveils an imbalance: only spin-down polarized electrons are allowed to be transferred from the semiconductor to the half-metal (spin-down) conduction band, while spin-up states remain blocked at the interface. This process determines an excess of spin-up states on the P3HT side, and due to a RKKY mechanism, this effective spin system becomes ferromagnetic polarized. Due to this proximity effect, the conjugate polymer becomes exchange coupled to the hard magnetic FePt (L10) phase, thus reducing the coercivity of the half-metal. These processes make this type of composite suitable for magnetic recording applications.
Article
Full-text available
The scientific and technological exploration of three-dimensional magnetic nanostructures is an emerging research field that opens the path to exciting novel physical phenomena, originating from the increased complexity in spin textures, topology, and frustration in three dimensions. One can also anticipate a tremendous potential for novel applications with those systems in a magnetic sensor and information processing technologies in terms of improved energy efficiency, processing speed, functionalities, and miniaturization of future spintronic devices. These three-dimensional structures are distinct from traditional bulk systems as they harness the scientific achievements of nanomagnetism, which aimed at lowering the dimensions down to the atomic scale, but expand those now in a tailored and designed way into the third dimension. This research update provides an overview of the scientific challenges and recent progress with regard to advances in synthesis approaches and state-of-the-art nanoscale characterization techniques that are prerequisite to understand, realize, and control the properties, behavior, and functionalities of three-dimensional magnetic nanostructures.
Chapter
We show that ferromagnetic resonance described for the first time in the middle of the twentieth century has found nowadays numerous applications in nanoscale magnetic systems. The article discusses how ferromagnetic resonance can be used to quantitatively extract magnetic key parameters such as magnetic anisotropy energy, interlayer exchange coupling, and the g-factor that provides access to orbital magnetism of single and coupled thin films as well as ensembles of magnetic nanoparticles. A short introduction to the theory and experimental approaches of ferromagnetic resonance is given in addition to the presentation of specific examples in detail.Keywords:nanoparticles;magnetism;magnetic anisotropy;ferromagnetic resonance;exchange coupling
Article
Owed to the large magneto-crystalline anisotropy (MCA) of the bulk FePt alloys, nanostructures with a few nm in diameter are considered for ultra-high density recording applications. First principles calculations in the framework of density functional theory (DFT) permit insight into the close interrelation between particle composition, morphology, and magnetism with access to the electronic level. The present survey will systematically highlight the impact of an additional encapsulation with Cu, Au, Al, and further main group elements on spin- and orbital magnetism and MCA with special emphasis on the role of the interface. Site resolved orbital moment anisotropy (OMA) of an uncovered 147 atom FePt nanoparticle.
Article
Full-text available
Some of the most relevant finite-size and surface effects in the magnetic and transport properties of magnetic fine particles and granular solids are reviewed. The stability of the particle magnetization, superparamagnetic regime and the magnetic relaxation are discussed. New phenomena appearing due to interparticle interactions, such as the collective state and non-equilibrium dynamics, are presented. Surface anisotropy and disorder, spin-wave excitations, as well as the enhancements of the coercive field and particle magnetization are also reviewed. The competition of surface and finite-size effects to settle the magnetic behaviour is addressed. Finally, two of the most relevant phenomena in the transport properties of granular solids are summarized namely, giant magnetoresistance in granular heterogeneous alloys and Coulomb gap in insulating granular solids.
Chapter
The magnetism of small particles or colloids has been investigated over many decades, and several reviews have dealt with the magnetic response of such particles ensembles [1, 2]. The sizes which can be prepared and are interesting for nanostructured materials range from a few atoms (“clusters”) to nanoparticles and colloids of a few micrometer diameter. All efforts on a detailed understanding of the magnetism of individual particles are hindered by the fact that particles prepared by either chemical synthesis, cluster beam techniques or Oswald ripening on single crystal surfaces are never perfectly alike. Hence, all magnetic measurements average over the properties of particles with different sizes , surfaces and shapes.
Article
Starting from an established correlation between the σ shape resonance position in nearedge x-ray-absorption fine-structure spectra and the intramolecular C-C bond length for gas-phase hydrocarbons, we derive a similar relation for chemisorbed molecules. The linear relation obtained allows us to determine the C-C bond length for chemisorbed molecules from near-edge x-ray-absorption fine-structure spectra with a ruler to an accuracy of <0.04 Å. Examples are given for bonding and bond-length changes of C2H2 and C2H4 on Pt(111).
Article
Magnetic force microscope (MFM) was used to characterize the L10 ordered FePt(001) films sputter deposited directly on MgO(001) substrates at an elevated temperature. With the change of nominal thickness (tN), the morphology varied from isolated particles to continuous films. The coercivity showed a marked change at the percolation boundary of tN≅45 nm, where the film morphology changed from a discontinuous to a continuous state. Below tN=45 nm, the coercivity did not change apparently, though the number of single-domain particles increased gradually with decreasing tN. At tN=20 nm, a critical (maximum) size of single domain particles, d=180 nm, was obtained from a size distribution, which was taken from the atomic force microscope/MFM measurement. The value calculated for this critical size was found to be d=155 nm in the assumption that the particles had ellipsoidal shape. The slight difference between experimental and theoretical values is likely to be attributed to an axis ratio (c/a) distribution of particles.
Article
The effect of surface anisotropy on the magnetic ground state of a ferromagnetic nanoparticle is investigated using atomic Monte Carlo simulation for spheres of radius R=6a and R=15a, where a is the interatomic spacing. It is found that the competition between surface and bulk magnetocrystalline anisotropy imposes a ``throttled'' spin structure where the spins of outer shells tend to orient normal to the surface while the core spins remain parallel to each other. For large values of surface anisotropy, the spins in sufficiently small particles become radially oriented either inward or outward in a ``hedgehog'' configuration with no net magnetization. Implications for FePt nanoparticles are discussed.
Article
We demonstrate that the long-range order parameter S of sputtered FePt (001) films may be improved by using postgrowth He ion irradiation. This was demonstrated both on disordered (S~0) and partially ordered (S~0.4) films in which S was increased up to 0.3 and 0.6, respectively. X-ray diffraction analysis showed that these changes are due to irradiation-induced chemical ordering. The changes in the magnetic hysteresis loops correlate with the expected perpendicular magnetic anisotropy increase. This method may find applications in ultrahigh-density magnetic recording.